-
公开(公告)号：EP0684716B1
公开(公告)日：2002-02-27
申请号：EP94480048.1
申请日：1994-05-25
-
2.发明公开A data communication network and method for operating said network 失效Datenübertragungsnetzund Verfahren zum Betrieben des Netzes。
公开(公告)号：EP0684716A1
公开(公告)日：1995-11-29
申请号：EP94480048.1
申请日：1994-05-25
摘要： A spanning tree organized architecture within a data communication network including a number of nodes interconnected by bi-directionnal links, wherein each said node is provided with means for dynamically setting and storing within the node a full topology database including full parent node relationship references. The system is also provided with means for fast path determination and fast spanning tree recovery, based on said topology databases contents.
摘要翻译： 在数据通信网络内的生成树组织架构，包括通过双向链路互连的多个节点，其中每个所述节点被提供有用于在节点内动态地设置和存储包括全部父节点关系参考的完整拓扑数据库的装置。 基于所述拓扑数据库内容，系统还提供快速路径确定和快速生成树恢复的方法。
-
3.发明公开Method and system for optimizing the end to end path selection in very large packet switching networks 审中-公开方法和系统，用于在非常大的分组交换网络优化端至端的路径选择
公开(公告)号：EP1009131A1
公开(公告)日：2000-06-14
申请号：EP98480090.4
申请日：1998-12-11
IPC分类号： H04L12/56
CPC分类号： H04L45/04 , H04L45/46 , H04L2012/562 , H04L2012/5621 , H04L2012/5687
摘要： The present invention relates to a method and system for optimizing the end to end path selection in very large packet switching networks.The network is divided in a backbone subnetwork and a plurality of access subnetworks. These subnetworks have a certain hierarchy, but direct links between access subnetworks are allowed. To fulfil the path request between nodes not pertaining to the same subnetwork, the path selection is performed in the backbone subnetwork by concatenating the backbone subnetwork topology with the access subnetwork topologies where the source or destination nodes are located. Only the required subset within access subnetwork topologies is transferred and cached into the backbone subnetwork topology. The main advantages are that :
The path is computed in one step only
There is no retry because there is no summarization.
The computation is faster because applied on a subset of the overall topology (the backbone subnetwork plus at most two access subnetworks).摘要翻译： 本发明涉及一种用于在非常大的分组交换网络networks.The优化端至端路径选择被分成骨架子网与接入子网的多个方法和系统。 这些子网有一定的层次，但接入子网之间的直接联系是允许的。 为了满足不属于相同的子网络节点之间的请求路径，路径选择是在主链中的子网络通过连接与所述接入子网拓扑，其中源节点或目的地节点位于骨干子网拓扑进行。 只有接入子网的拓扑结构中所需要的子集，转移和高速缓存到骨干子网的拓扑结构。 的主要优点是：路径计算在一个步骤中仅没有重试因为没有总结。 计算速度更快，因为对整体拓扑结构（主干子网加最多两个接入子网）的一个子集应用。
-
4.发明公开Distributed directory services for locating network resources in a very large packet switching network 审中-公开在一个非常大的分组交换网络中定位网络资源的分布式目录服务
公开(公告)号：EP1009130A1
公开(公告)日：2000-06-14
申请号：EP98480089.6
申请日：1998-12-11
IPC分类号： H04L12/56
CPC分类号： H04L45/04 , H04L45/46 , H04L2012/562 , H04L2012/5621 , H04L2012/5687
摘要： The present invention relates to a method and system of locating a network resource in a packet switching network divided in a backbone subnetwork connected to one or a plurality of access subnetworks. The access subnetworks comprise one or a plurality of access border nodes directly connected to the backbone subnetwork. The backbone subnetwork comprises one or a plurality of backbone border nodes directly connected to the access subnetworks. Border nodes of different subnetworks are connected by border links.
The method in access border nodes involves the steps of :
receiving a query message comprising a resource name, the resource name including a prefix;
comparing the resource name prefix with the prefixes stored in an access border node directory database, said access border node directory database comprising prefixes of the resources local to the access border node subnetwork;
determining whether the resource name relates to a resource which is local to the access border node subnetwork or not;
forwarding said query message to an adjacent backbone border node when the resource name relates to a resource which is not local to the access border node subnetwork.
The method in backbone border nodes involves the steps of :
receiving a query message comprising a resource name, the resource name including a prefix;
comparing the resource name prefix with the prefixes stored in one or a plurality of backbone border node directory databases. Each backbone border node directory database corresponds to an access subnetwork managed by the backbone border node;
determining whether the resource name relates to a resource which belongs to an access subnetwork that the backbone border node manages or not;
forwarding the query message to an adjacent access border node when the resource name relates to a resource which belongs to an access border node subnetwork managed by the backbone border.
Each access subnetwork has its own spanning tree and distributed topology, and interactions between access networks are limited to a minimum. The size of distributed topologies is kept small and the control traffic is mostly limited to subnetworks.摘要翻译： 本发明涉及定位在分组交换网络划分成连接到一个或接入子网的多个骨干子网的网络资源的方法和系统。 访问子网包括一个或接入边界的多个节点直接连接到主干网子网。 骨干子网包含一个或骨干边境的多个节点直接连接到接入子网。 不同子网的边界节点通过边界左连接。 在接入节点边界的方法，包括以下步骤：接收查询消息包括资源名称，资源名称包括前缀; 比较资源名称前缀与存储在访问边界节点目录数据库的前缀，所述接入边界节点目录数据库，包括资源的前缀本地接入节点边界子网; 确定性矿山资源是否名涉及资源全部是本地的接入节点子网边界或没有; 说在邻近骨架边界节点转发查询消息当资源名称涉及资源全部是不是本地的接入边界节点子网。 在骨干节点边框的方法包括以下步骤：接收查询消息包括资源名称，资源名称包括前缀; 比较资源名称前缀存储在一个前缀或骨干边界节点目录数据库的多元性。 每个骨干边界节点目录数据库对应于由骨干边界节点管理接入子网; 确定性矿山资源是否名涉及属于接入子网做的骨干边界节点管理或不是资源; 在相邻接入边界节点转发查询消息当资源名称涉及属于能够访问由骨干边境管理边界节点子网的资源。 每个接入子网都有自己的生成树拓扑结构和分布，以及接入网络之间的交互限制到最小。 分布式拓扑的尺寸保持较小和控制业务大多局限于子网。
-
5.发明公开Method for preempting connections in high speed packet switching networks 失效在einem Hochgeschwindigkeits-Paketvermittlungsnetz的Verfahren zur Vorrangunterbrechung von Verbindungen
公开(公告)号：EP0714192A1
公开(公告)日：1996-05-29
申请号：EP94480143.0
申请日：1994-11-24
IPC分类号： H04L12/56
CPC分类号： H04L12/5602 , H04L2012/5632 , H04L2012/5651
摘要： The present invention deals with the capability of high speed network nodes to establish new connections by preempting existing ones when insufficient bandwidth is available in the loaded network. More particularly, it discloses the arrangement of numerous preemption priorities in a limited number of groups and the definition of a median priority level within each group. This median, stored in the topology database of each node of the network allows a better bandwidth utilization per group while consuming very little memory space and generating very low traffic overhead.
摘要翻译： 本发明涉及高速网络节点通过在加载的网络中不足带宽可用时抢占现有网络来建立新连接的能力。 更具体地，其公开了在有限数量的组中的许多抢占优先级的排列以及每个组内的中间优先级的定义。 存储在网络每个节点的拓扑数据库中的这个中值允许每组更好的带宽利用率，同时消耗很少的存储空间并且产生非常低的业务开销。
-
-
-
-
搜索结果
5 条记录 (耗时 0.013 秒)
